Retroviral infections take an enormous toll on human health. The human immunodeficiency virus type 1 (HIV- 1, or HIV) has killed 30 million people worldwide and 36 million people are living with HIV/AIDS. There is no effective vaccine for HIV. The available antiretroviral therapies (ARTs) for treating HIV cannot cure infected patients. ART must be taken life-long because HIV can exist in a dormant state by latently infecting CD4+ T cells. These latent reservoirs are long lived, ensuring lifelong persistence of the virus, and are recognized as the greatest obstacle to eradicating HIV from patients. Approaches to 'activate and kill' these latent reservoirs, and cure HIV-infected individuals, are being actively pursued. However, even under ideal laboratory conditions, the most powerful activators only partially reactivate latent HIV. We have established that this heterogeneity results in large part from stochastic fluctuations in transcription that drive a fate 'switch' in HIV. If we hope to efficienty reactivate latent HIV, it is critical to characterize the molecular mechanisms driving these transcriptional fluctuations and address how the 'switch' between active and latent infection is regulated. Our long-term goal is to identify the molecular pathways to efficiently 'activate and kil' latent HIV. The objectives of this project are to develop a quantitative model of HIV latency, experimentally validate this model in donor-derived primary CD4+ T cells, and perturb the sources of variability that generate partial reactivation of latent HIV. Based upon our extensive preliminary studies, our central hypothesis is that stochastic fluctuations in HIV transcription (ie. 'noise') limit HIV reactivation and that manipulating noise will enhance HIV reactivation. In bacteria and phage, tuning gene-expression variability can significantly alter similar cell-fate decisions. The rationale for this project is that identifying approaches to tune HIV variability wil enable us to tune HIV latent reactivation and efficiently purge of the latent reservoir. We will achieve our objective through specific aims that rely on single-cell imaging and mathematical modeling of single-cell data. Specifically, we capitalize on a new suite of microwell devices and imaging approaches to develop a mathematical model of HIV latency in donor-derived primary CD4+ T cells. We will identify the molecular sources of stochastic fluctuations to determine which parameters are most sensitive to perturbation. This model will enable us to rationally test new approaches for reactivating latent HIV in primary CD4+ T cells. In addition to the medical relevance, the proposed research has broad significance since the mechanisms driving variability in fate-decision switches are unclear in general, especially in mammalian systems. This project would provide a much-needed quantitative characterization of a noise-driven developmental switch in a mammalian system. Ultimately, the knowledge gained will guide new approaches to tune fate switches not just in HIV, but also in diverse mammalian systems.